Robotic grinding is a promising automatic technique for free-form surface manufacturing. One important problem that restrains the application of robotic grinding is the machining quality. Existing methods considers only individual kinematic errors or joint stiffness. In this article, a systematic method of error compensation, workpiece position optimization and tool pose optimization is proposed to reduce the machining error. First, the mathematical models of machining error with respect to both kinematic errors and joint stiffness are built using speed and force adjoint transformation, respectively. Compared with existing indirect index evaluation models, this model is improved by directly building the quantitative function relationship between the machining error and the corresponding factors. Based on the model, an error compensation strategy is presented by only fine-tuning the workpiece frame position. Then, an objective function based on the compensated machining error is defined to optimize both the workpiece position and the tool pose. Experiments demonstrate the availability of the proposed method for machining error reduction.

中文翻译：

---

减少机器人磨削加工误差的系统模型

机器人磨削是一种有前途的自动化技术，用于自由曲面的制造。限制机器人磨削应用的一个重要问题是加工质量。现有方法仅考虑单个运动学误差或关节刚度。本文提出了一种误差补偿，工件位置优化和刀具姿态优化的系统方法，以减少加工误差。首先，分别使用速度和力伴随变换建立关于运动学误差和关节刚度的加工误差的数学模型。与现有的间接指标评估模型相比，该模型通过直接建立加工误差与相应因素之间的定量函数关系进行了改进。根据该模型，仅通过微调工件框架的位置即可提出误差补偿策略。然后，基于补偿的加工误差定义目标函数，以优化工件位置和刀具姿态。实验证明了所提出的减少加工误差的方法的可行性。